Compared to younger adults, older adults not only have difficulty remembering the content of information, but they also have problems in remembering the source (e.g., speaker, temporal or spatial context, modality) of information. The present project examines age-related differences in making source monitoring judgments and in remembering source-specifying attributes. The work is guided by a source monitoring framework (Johnson, Hastroudi, & Lindsay, 1993) and general memory model (Johnson, 1992; Johnson & Chalfonte, 1994; Johnson & Hirst, 1993) that identify relevant features of complex memories and describe central encoding and source monitoring processes. The proposed experiments assess the relative availability of certain features of memories, potential age differences in the binding of features into complex memories, and the contribution of different types of remembered information (e.g., perceptual, contextual and cognitive operations) to age differences in remembering source. We also examine whether age deficits in remembering source are dissociated from deficits in remembering content, and assess the impact of variables (e.g., repetition, specific combinations of features to be bound, amount of potential confusion among features, delay) on memory for content and source. Finally, we investigate the relations between performance on source memory tasks and neuropsychological tests sensitive to dysfunction in frontal, medial-temporal, and parietal brain regions. The findings from these experiments will help characterize the nature of age- related differences in forming and remembering complex events. Furthermore, the data and theoretical ideas generated from the empirical studies will provide constraints for developing a computer simulation model of age differences in encoding, reviving and making source judgments about complex memories. Devising assessment procedures for identifying and quantifying source monitoring deficits, correlating them with other indices (such as tests for frontal system deficits), and developing clinical remedial techniques for dealing with such deficits, will benefit from a more specific characterization of source monitoring situations and relevant memory attributes and processes.